Application of the Shear Band Propagation Method to Slope Stability Analysis of a Palaeo-Landslide in the Caspian Sea

Abstract

Geophysical observations of palaeolandslides in offshore settings include long and relatively thin failures that occur on very mild slopes. Back analysis of such failures using the limiting equilibrium approach and measured geotechnical soil properties is typically used to inform the likely causes of past failures. An assessment can then be made of the present day significance of causal factors (e.g. the presence of excess pore water pressure), with the limiting equilibrium approach used to assess factors of safety for present-day slopes and predict the potential size of any future failures. However, for long, thin slides, use of the limiting equilibrium approach can lead to the conclusion that either a significantly large seismic trigger or other, abnormal geological condition was required to initiate failure. A result has been the focusing of academic research into other failure causes and conditioning factors. Further, the stability of the present day seabed for offshore developments can be overestimated, given the rarity of large seismic events in many areas and geotechnical measurements that not always show abnormal soil conditions. The shear band propagation approach offers an alternative view, where an initial small failure can grow into a larger failure at the point of initiation. A result is that an explanation of the conditions leading to failure can be attributed to many submarine landslides. Further, accounting for the potential propagation of a shear band in the future prediction of submarine landslides can result in greater predicted failure sizes and lower factors of safety than using limiting equilibrium only. The application of the shear band propagation approach to back analyse a palaeolandslide in the ACG Field, Caspian Sea, is used to demonstrate the methodology and compare with results using the limiting equilibrium approach. Results show that the shear band propagation approach successfully predicts observed failure geometries and suggests triggering conditions from a lower magnitude seismic event compared to that predicted using limiting equilibrium. The future application of shear band propagation theory to submarine landslide analysis is also discussed.

Geotechnical Challenges in the Caspian Sea

“…A novel approach to slope stability was adopted as described in Gray et al 2015 andRushton et al 2015. Once the probability of landside occurrence was refined using the combination of historic frequency and future probability prediction, the spatial probability of a given event hitting a given piece of infrastructure was assessed, and finally the vulnerability of the equipment to a hit was assessed to derive the Top Event probability (as shown in map form in Figure 10). A novel approach to slope stability was adopted as described in Gray et al 2015 andRushton et al 2015. Once the probability of landside occurrence was refined using the combination of historic frequency and future probability prediction, the spatial probability of a given event hitting a given piece of infrastructure was assessed, and finally the vulnerability of the equipment to a hit was assessed to derive the Top Event probability (as shown in map form in Figure 10).…”

“…The methodology is described in more detail in Puzrin and Germanovich, 2005;Puzrin et al, 2015 andGray et al 2015 but essentially it is an approach through a readily determinable parameter (the gravitational shear stress ratio) which can be applied to the upslope and downslope propagation of a shear band initiated on a locally steep slope. Each approach has its benefits and also shortcomings, but combining them in the right way should yield the most reliable result.…”

Simple criteria for ploughing and runout in post-failure evolution of submarine landslides

“…The most recent sediments comprise soft clay draped over an older "mega" landslide (MM700). A complex of landslides that occurred about 5 ka BP within this drape (MM900) has been described by Gray et al (2015). At the top of the failure scarp, MM900 does not exhibit any traces of the spreading failure (Fig.…”

“…Substituting soil properties from Table 1 together with geometric parameters from Table 2 (after Gray et al 2015) into eqs. (31), (40), and (41) provides confirmation that after an earthquake the bottom of the MM900 failure scarp in Fig.…”

“…6 has been explained by Puzrin et al (2016b) using SBP spreading failure criteria with parameters from Table 1 given in Gray et al (2015). The maximum thickness of the sediments involved in the MM900 mass movement was h z = 25 m. It has been shown that, on one hand, for the spreading failure not to occur for MM900 slab failures with h z ≤ 25 m and, on the other hand, for it to occur in the same cross section in the much deeper and older landslide MM700, the characteristic displacement has to stay within a rather narrow interval of ␦ = 0.10-0.16 m.…”